I remember 20 years ago or so I encountered a similar problem, but I forgot
the solution. I opened the classical paper of
J.-P. Boy, M. Llubes, Jacques Hinderer, and Nicolas Florsch,
"A comparison of tidal ocean loading models using superconducting gravimeter
data", Journal of Geophysical Research, vol. 108, NO. B4, 2193,
doi:10.1029/2002JB002050, 2003 and found there:
Aga! My memory is still working. But the phase plot is not consistent with this explanation. Nothing surprising, since FES2014b appeared ten years after the paper was published.
Then I just added 180° to the phase and computed SSa ocean loading:
As a control example, I computed the equilibrium ocean loading
for the SSa. The dynamic SSa tide should be different from the equilibrium
tide, but the differences are expected not to be so significant to
wash out the phase pattern around ±35.27°. Here is the result:
The equilibrium SSa ocean loading looks similar to the FES2014b SSa
ocean loading computed with adding 180° phase everywhere,
but not only at |φ| > 35.27° as J.-P. Boy, et al. (2003)
recommend.
The equilibrium loading looks reasonable. The SSa tide-generation potential is Ampl * ( 1.5 sin φ2 - 0.5 ) * cos ( ω t + ψ ), where φ is geocentric latitude. In the polar regions ( 1.5 sin φ2 - 0.5 ) > 0 and the tide-generating potential is positive. Therefore, the ocean tide is positive. Therefore, the ocean loading is negative. In the equatorial regions ( 1.5 sin φ2 - 0.5 ) < 0 and the ocean loading is positive.
Still puzzles me why 180° phase shift did occur in the ocean tide models in the first place? Somebody long time ago made a mistake, and this mistake became unwritten "convention"?
Last update: 2017.06.15_10:46:38
Web page was prepared by Leonid Petrov
(
)